Multimodal speech systems
The paper describes the various knowledge sources required to handle human-machine multimodal interaction efficiently: these sources constitute the task, user, dialogue, environment and system models. The first part of the paper discusses the content of these models, emphasizing the problems occurring when speech is combined with other modalities (e.g. vision, hearing, taste, gesture, haptic properties etc. and their various forms). The second part of the paper focuses on spoken language characteristics, describes different parsing methods (mainly rule-based and stochastic approaches) using a task model, and presents the integration of the rule-based method in an end-to-end information retrieval system. The results obtained with the rule-based and stochastic methods used for representing the task model show that stochastic approach slightly outperforms the rule-based method. Furthermore, such a stochastic method should be more portable with respect to different languages and similar task domains. For the entire collection see [Zbl 0927.00044].
This page was built for publication: Multimodal speech systems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4942762)
